A Selective ε-Protein Kinase C Antagonist Inhibits Protection of Cardiac Myocytes from Hypoxia-induced Cell Death*

Protein kinase C activation is thought to protect cardiac tissue from subsequent ischemic injury by a process termed preconditioning. The protein kinase C isozyme that mediates preconditioning has not yet been identified. Using a cell culture model of hypoxic preconditioning, we found that cardiac myocyte viability after 9 h of hypoxia was increased by more than 50% over control. Preconditioning activated protein kinase C isozymes as evidenced by translocation from one cell compartment to another as follows: there was a 2.1-fold increase in ε-protein kinase C activation, a 2.8-fold increase in δ-protein kinase C activation, and no increase in βI-protein kinase C activation. 4β-Phorbol 12-myristate 13-acetate mimicked hypoxic preconditioning, increasing myocyte survival after prolonged hypoxia by 34% compared with control. We previously identified an ε-protein kinase C-selective antagonist, εV1-2 peptide, that inhibits ε-protein kinase C translocation and function in cardiac myocytes (Johnson, J. A., Gray, M. O., Chen, C.-H., and Mochly-Rosen, D. (1996) J. Biol. Chem. 271, 24962–24966). εV1-2 peptide abolished hypoxic preconditioning and phorbol ester-mediated cardiac protection. Therefore, preconditioning can be induced in this culture model, and activation of ε-protein kinase C is critical for cardiac myocyte protection.

Identification of novel therapeutic targets for the prevention of ischemia-induced cardiac injury has been an area of intense investigation for the past 10 years. First described in a canine heart model (1), preconditioning of cardiac tissue with one or more brief episodes of ischemia remains one of the most potent experimental means of reducing irreversible tissue injury during subsequent prolonged ischemia. However, the cellular and molecular mechanisms underlying this protective phenomenon remain obscure. Numerous molecules such as adenosine, ␣ 1agonists, angiotensin II, bradykinin, and opioids have been invoked as potential mediators of ischemic preconditioning in whole heart models (2)(3)(4)(5)(6).
Although the exogenous mediators of preconditioning have not been definitively identified, there is substantial experimental evidence that one of the major intracellular signal trans-duction pathways underlying cardiac protection involves activation of protein kinase C (PKC). 1 At least six different PKC isozymes have been identified in rat cardiac myocytes (7,8), where PKC regulates a number of functions such as force of contraction (9), atrial natriuretic factor secretion (10), and gene expression (11). Individual isozymes translocate to characteristic intracellular sites following activation (12,13) with each isozyme playing a different role in myocyte function (14,15). Activation of PKC correlates closely with the cardiac protection mediated by ischemic preconditioning in rat whole heart models. For example, PKC stimulation with dioctanoylglycerol protected adult rat heart during 45 min of regional ischemia, whereas PKC inhibition with chelerythrine abolished protection mediated by ischemic preconditioning (16). In addition, activation of PKC by transient ischemia or ␣ 1 -agonists in an isolated rat heart model induced protection against global ischemia/reperfusion injury that was inhibited by PKC antagonists (17).
Studies of ischemic preconditioning in whole heart models have been limited by technical issues such as the relatively low specificity of available antagonists for PKC versus other kinases, the inability of agonists and antagonists to discriminate among multiple PKC isozymes, and the difficulty of examining signal transduction mechanisms at the level of the individual cell. Cardiac myocyte culture models may provide complementary approaches to the investigation of signal transduction in preconditioning. In one such culture model, a 25-min exposure to hypoxia followed by reoxygenation protected cardiac myocytes against membrane damage for up to 6 h of severe hypoxia, and pretreatment of cells with phorbol ester mimicked the protective effects of hypoxic preconditioning (18).
In this study, we identified two hypoxic preconditioning protocols that substantially improved the viability of cultured neonatal rat cardiac myocytes following several hours of profound hypoxia. We found that both protocols induced a selective activation of PKC isozymes as evidenced by translocation and that direct activation of PKC with a phorbol ester mimicked the protective effect of hypoxic preconditioning. Moreover, we used a novel 8-amino acid antagonist of ⑀PKC, ⑀V1-2 peptide, to identify the isozyme that mediates this protective effect. We previously proposed the use in general of such isozyme-selective peptide inhibitors of PKC translocation and function (19). ⑀V1-2 peptide is derived from the first unique region (V1) of ⑀PKC (amino acids 14 -21), and its action as a selective antagonist of ⑀PKC translocation and function in cardiac myocytes has been characterized in detail (20). Introduction of this peptide into myocytes selectively inhibited PMA-induced translocation of ⑀PKC but not the translocation of other PKC isozymes (20). Furthermore, there was a selective loss of regulation of * This study was supported in part by NHLBI Grants HL-52141 (to D. M.-R.) and HL-25847 (Project 3, to J. S. K.) from the National Institutes of Health and by the Research Service of the Department of Veterans Affairs (to J. S. K.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
¶ contraction rate in the presence of ⑀V1-2 that was not seen with ␤C2-4, a translocation inhibitor selective for the cPKC isozymes (20,21). Using ⑀V1-2 in a different cell culture model, we also showed that ⑀PKC mediates, at least in part, glucoseinduced insulin secretion in pancreatic cells (22). In the present study, we introduced ⑀V1-2 peptide into cardiac myocytes and showed for the first time that isozyme-selective inhibition of ⑀PKC translocation and function inhibited the protective effect of preconditioning. These data indicate a major role for the ⑀PKC isozyme in cardiac myocyte protection by hypoxic preconditioning.

EXPERIMENTAL PROCEDURES
Ventricular Myocyte Preparation-Primary cultures were prepared as described previously (23). After removal of non-myocytes in a preplating step, isolated myocytes (90 -95% of the cells) were seeded at 800 cells per mm 2 into 8-well glass chamber slides (Nunc) for cell viability assays and immunofluorescence staining, into 6-well plastic culture plates (Fisher) for lactate dehydrogenase (LDH) assays, or into 100-mm glass culture dishes for Western blot analysis. On culture day 4, myocytes were placed in defined medium (23) then fed with fresh defined medium on the evening prior to treatment (days 6 -8).
Induction of Hypoxia-For hypoxic preconditioning and prolonged hypoxic challenges, myocytes were transferred into and out of a plexiglass glove box (Anaerobic Systems) maintained at 37°C with a humidified atmosphere of 1% CO 2 , less than 0.5% O 2 , and the balance N 2 . Monitoring with a Fyrite Gas Analyzer (United Technologies) verified O 2 concentrations between 0.2 and 0.5%. Lysis of samples for Western blot analyses and fixation of myocytes for immunofluorescence staining were performed without reoxygenation. Normoxic incubations of myocytes were conducted in a water-jacketed incubator (Forma Scientific) gassed with 99% air and 1% CO 2 at 37°C.
Hypoxic preconditioning initially consisted of four 90-min periods of hypoxia alternating with 60-min normoxic incubations. Myocytes were kept in 5 mM glucose defined medium throughout the protocol. For subsequent single cycle preconditioning experiments, myocytes were fed pre-equilibrated, glucose-free defined medium within the hypoxia chamber for 30 min. For phorbol ester preconditioning experiments, myocytes were stimulated with 10 nM PMA for 10 min then washed with fresh 5 mM glucose defined medium.
We previously reported that 2 h of hypoxia resulted in a medium pO 2 of 23.9 Ϯ 1.5 torr, pH of 7.34 Ϯ 0.02, and pCO 2 of 46.6 Ϯ 0.7 torr. LDH release from glucose-supplemented myocytes subjected to 2 h of hypoxia and from glucose-deprived myocytes subjected to 30 or 60 min of hypoxia was no different from normoxic control cells (24). These data suggested that hypoxic preconditioning does not cause irreversible cellular damage, an assumption that we verified using a cell viability assay (see below). In all cases, myocytes were subjected to a prolonged hypoxic challenge by transfer into the plexiglass chamber followed by feeding with glucose-free defined medium pre-equilibrated for several hours within the hypoxic environment. Cells were removed 7-9 h later and immediately underwent the indicated analysis.
Western Blot Analysis-Western blot analysis was carried out as described previously in our laboratory (14) using monoclonal anti-␣PKC and anti-␤PKC antibodies (Seikagaku) diluted 1:1000 and then rabbit anti-mouse IgG antibodies (Cappel) at 1:1000 dilution or using rabbit polyclonal anti-␦PKC and anti-⑀PKC antibodies (Life Technologies, Inc.) at 1:300 dilution, followed by 125 I-protein A (ICN) and detection of PKC immunoreactive bands by autoradiography.
Cell Viability Assay-Following treatment of myocytes grown on chamber slides, living and dead cells were distinguished using the Eukolight TM Viability/Cytotoxicity assay (Molecular Probes). Culture medium was replaced with 2 M calcein acetoxymethyl ester and 4 M ethidium homodimer-1. Slides were viewed with a Zeiss IM35 microscope using a 40 ϫ water immersion objective. Viable (green fluorescent by calcein) and non-viable (red fluorescent by ethidium) myocytes present in 10 -20 random microscopic fields per condition per experiment were recorded.
Immunofluorescence Staining-Immunofluorescence staining was carried out using identical antibodies as described previously (8,20). Isozyme-specific rabbit polyclonal antibodies (R & D Antibodies) directed against ␤ I PKC, ␦PKC, or ⑀PKC were used at 1:100 dilution followed by fluorescein isothiocyanate-conjugated goat anti-rabbit IgG antibodies (Cappel) diluted 1:1000 and were viewed with a Zeiss IM35 microscope using a 40 ϫ water immersion objective. Images were recorded on Kodak TMAX 400 film with exposure time of 1 s for all photomicrographs and were processed by an automated, commercial developer without additional adjustment. The specificity of the immunostaining technique used to detect ⑀PKC in the present study has been confirmed previously. First, pre-absorption of primary antibody with the immunizing peptide (8) or with the corresponding recombinant PKC expressed in Sf9 cells (20) abolished specific staining (see also Fig. 3C). Second, there was a direct correlation between translocation of ⑀PKC measured by Western blot analysis and that measured by immunofluoresence staining (14,20,21). Finally, the immunostaining pattern seen with commercial antibody was identical to that found using monoclonal antibody (CK 1.4) to ⑀PKC (25).
Lactate Dehydrogenase Assay-The effect of hypoxia on LDH release from cardiac myocytes was assessed as described previously (24). Following each treatment, culture media were stored at 4°C. An equal volume of cold lysing buffer (10 mM Tris-HCl, pH 7.4, and 1 mM EDTA) was added to each well, and lysates were centrifuged at 4°C for 15 min at 50,000 ϫ g. Aliquots from culture media samples (released LDH) and from supernatants of cell lysates (retained LDH) were analyzed using an assay in which the rate of decrease in absorbance of the sample at 340 nm is directly proportional to LDH activity (Sigma). LDH activity was expressed as units per liter, where 1 unit is defined as the amount of enzyme catalyzing the formation of 1 mol/liter NAD per min under the conditions of the assay.
Transient Permeabilization of Cardiac Myocytes-Transient permeabilization of cardiac myocytes was carried out exactly as described previously using buffer containing saponin (50 g/ml) and ATP (6 mM) at 4°C (26). Permeabilization carried out according to this protocol did not alter myocyte viability, spontaneous or stimulated contraction rates, basal or hormone-induced expression of c-fos mRNA, or growth factor-induced hypertrophy (26).
Statistical Analysis-All data were expressed as mean Ϯ S.E. Numerical data were compared using Student's t test for paired observations between two groups and by analysis of variance followed by the Bonferroni t test when more than two groups were analyzed. A p value of Ͻ0.05 was considered significant.

Characteristics of the Hypoxic Preconditioning Model-Our
original preconditioning paradigm consisted of four 90-min periods of hypoxia alternating with 60-min normoxic incubations. The hypoxic period was based on our observation that myocytes incubated in glucose-supplemented medium tolerated up to 2 h of hypoxia without reduction of viability (24). The number of cycles was based on the study of preconditioning in canine myocardium in which four brief coronary artery occlusions protected cardiac tissue from prolonged ischemia (1). The protective effect of our protocol was determined using an assay that discriminated between living and dead myocytes upon completion of a prolonged hypoxic challenge. This method has been validated with a variety of adherent cell types (27) as well as with primary cultures of neonatal rat cardiac myocytes (28).
The proportion of viable (green fluorescent) cardiac myocytes under normoxic conditions was consistently greater than 95% (Fig. 1A). Following transfer, non-viable (red fluorescent) cardiac myocytes exceeded 50% after 7-9 h of exposure to the hypoxic environment (Fig. 1B). In two independent experiments, 20 random microscopic fields per condition (600 cells each) were scored just after the final preconditioning cycle. The proportion of viable cells in the preconditioned group was no different from that of normoxic control cells (96.5 Ϯ 0.7 versus 97.2 Ϯ 0.8%). Therefore, hypoxic preconditioning had no immediate effect on myocyte survival.
In contrast, hypoxic preconditioning had a profound effect on expression of the inducible 72-kDa member of the heat shock protein 70-kDa family (Hsp70). In control myocytes Hsp70 was barely detectable by Western blot analysis, whereas precondi-tioning up-regulated the expression of Hsp70 by severalfold ( Fig. 2A). Others (29) have reported that overexpression of rat Hsp70 improved contractile function and reduced the zone of infarction following 20 min of zero-flow ischemia in hearts harvested from transgene-positive mice compared with hearts from transgene-negative littermates. Also, specific induction of Hsp70 in cultured neonatal rat cardiac myocytes by the tyrosine kinase inhibitor herbimycin-A increased cell survival during subsequent lethal heat stress and simulated ischemia (30). We found that preconditioning increased the proportion of surviving myocytes by 52% compared with control (66.5 Ϯ 1.4 versus 43.7 Ϯ 3.3%) following 9 h of hypoxia (Fig. 2B). Therefore, we have identified a cardiac myocyte culture model for hypoxic preconditioning.
Involvement of PKC in Preconditioning-We first determined whether hypoxic preconditioning activated PKC using an immunofluorescence assay in which myocytes were stained with isozyme-specific anti-PKC antibodies. Activated PKC isozymes translocate to distinct, characteristic subcellular locations; ⑀PKC translocates from the nucleus to perinuclear, cell-cell contact, and cross-striated structures. ␦PKC translocates from the nucleus to perinuclear structures. ␤ I PKC translocates from the cytosol into the nucleus (8). Individual myocytes can then be scored as basal or activated, and the proportions of each were compared with those of cell populations stimulated with known activators of PKC such as norepinephrine. This immunofluorescence technique is extremely sensitive in demonstrating translocation because it does not disrupt cell structure and because dissociation from the translocation site does not occur in cells fixed immediately after treatment. We found that hypoxic preconditioning produced a 2.1-fold increase in ⑀PKC activation (49.9 Ϯ 2.8 versus 23.6 Ϯ 2.6% for control) and a 2.8-fold increase in ␦PKC activation (58.4 Ϯ 2.9 versus 20.7 Ϯ 2.8% for control), an effect comparable to that induced by stimulation with 2 M norepinephrine under normoxic conditions. No activation of ␤ I PKC by hypoxic preconditioning was observed (Fig. 3). Therefore, hypoxic preconditioning selectively activated ␦and ⑀PKC isozymes in cultured cardiac myocytes. A recent study using a similar myocyte culture model reported redistribution of ⑀PKC in response to hypoxia using Western blot analysis (31), thereby providing independent corroboration of our observations.
We also determined whether direct activation of PKC with phorbol ester could protect cardiac myocytes from hypoxic damage. We stimulated myocytes with 10 nM PMA for 10 min since this concentration and incubation period induced a robust, selective activation of ␦and ⑀PKC by Western blot analysis (Fig. 4). As we previously reported, there was no activation of ␣PKC and minimal activation of ␤PKC using Ͻ10 nM PMA (14). In contrast, where activation of these isozymes has been observed, the concentration of PMA used was 100 nM to 1 M (7, 32, 33). This pattern of PKC activation by 10 nM PMA was comparable to that observed with hypoxic preconditioning as determined by immunofluorescence staining (Fig. 3). Furthermore, PMA-mediated preconditioning protected cardiac myocytes, increasing the proportion of surviving myocytes by 34% compared with control (59.3 Ϯ 1.9 versus 44.1 Ϯ 2.1%) following 9 h of hypoxia (Fig. 5A). PMA-mediated preconditioning was also protective during prolonged hypoxia by an independent assay of cell viability and retained LDH activity. Under normoxic conditions PMA had no effect on retained LDH activity (data not shown). However, PMA increased retained LDH activity by 42% compared with control (387 Ϯ 18 units/liter versus 274 Ϯ 24 units/liter in four independent experiments, p Ͻ 0.05) following 9 h of hypoxia. Therefore, PMA treatment of myocytes in culture mimicked hypoxic preconditioning.
Use of an Isozyme-selective ⑀PKC Peptide Antagonist-In many cell types, activation of PKC isozymes is associated with translocation from the cell soluble to the cell particulate fraction (34). In addition to binding to lipids, activated PKC isozymes bind to specific anchoring proteins within the particulate fraction termed RACKs, for receptors for activated protein kinase C (35). If stimulation-induced translocation of a PKC isozyme to its appropriate intracellular target is required for its function, then peptides that compete for binding of activated PKC isozymes to their RACKs should act as isozymeselective inhibitors of PKC translocation and function. For example, a 9-amino acid peptide derived from the second common region (C2) of ␤PKC, termed ␤C2-4 (SLNPEWNET, ␤PKC-(218 -226)), selectively inhibited insulin-induced ␤PKC translocation and function in Xenopus oocytes as well as PMAmediated ␤PKC translocation in cultured neonatal rat cardiac myocytes (21). In a more recent study, the first variable region (V1) of ⑀PKC selectively inhibited PMA-induced ⑀PKC translocation and function in cultured neonatal rat cardiac myocytes. An 8-amino acid peptide derived from ⑀V1, termed ⑀V1-2 (EAVSLKPT, ⑀PKC-(14 -21)), displayed almost identical properties as an antagonist of ⑀PKC translocation and function (20). Since ⑀PKC is activated by preconditioning, we used this isozyme-selective peptide antagonist to examine the role of ⑀PKC in the signal transduction pathway underlying protecion against hypoxic injury.
We used a transient, saponin-based permeabilization method developed in our laboratory to introduce ⑀V1-2 and control peptides into cells. This technique has been characterized extensively in cultured neonatal rat cardiac myocytes and does not compromise cell viability, spontaneous or stimulated contraction rates, basal or hormone-induced expression of c-fos mRNA, or growth factor-induced hypertrophy (26). We found that the protective effect of PMA-mediated preconditioning was completely abolished by ⑀V1-2 and unaffected by control peptides such as ⑀V1-3 (Fig. 5A). (Note that ⑀V1-3 is a peptide derived from the ⑀PKC V1 region that does not act as an ⑀PKC translocation inhibitor (20).) Therefore, ⑀PKC mediates, at least in part, PMA-induced preconditioning in cultured cardiac myocytes.
Because activation of ⑀PKC appeared to be critical to the protective effect of PMA-mediated preconditioning, we optimized our hypoxic preconditioning protocol by using ⑀PKC translocation as a marker of cardiac protection. PKC translocation has traditionally been monitored by Western blot analysis, a method that may be less sensitive than immunofluores- cence staining because it depends on irreversible association of PKC isozymes with the particulate fraction. Nevertheless, when we used Western blot analysis to determine PKC translocation, we found a robust activation of ⑀PKC in myocytes maintained in glucose-free defined medium and incubated under hypoxic conditions for a single cycle lasting 30 min or longer (Fig. 5C, left). Furthermore, hypoxia-induced translocation of ⑀PKC from the soluble to the particulate fraction was inhibited by the presence of ⑀V1-2 (Fig. 5C, right). The total amount of ⑀PKC present in cell fractions from cardiac myocytes following activation by hypoxia was diminished in the presence of ⑀V1-2. Our recent study with Messing and collaborators (36) showed that stable expression of the ⑀V1 fragment in PC12 cells caused a similar decline in ⑀PKC levels, observed only after PMA activation but not in unstimulated cells. A 5-min incubation with 30 nM PMA resulted in loss of ⑀PKC from the soluble fraction in both control and ⑀V1 fragment-containing cells. In control cells there was a corresponding increase in the amount of ⑀PKC in the particulate fraction indicating translocation of the enzyme, whereas no increase was seen in cells expressing ⑀V1 fragment (36). Taken together with the data presented in the current study, these findings suggest that inhibition of translocation of activated ⑀PKC resulted in increased sensitivity of the enzyme to degradation.
Finally, we determined whether 30 min of hypoxia in the absence of glucose protected cardiac myocytes during subsequent prolonged hypoxia and whether ⑀V1-2 inhibited this protection. A 30-min hypoxic period was chosen for preconditioning because it was both long enough to induce ⑀PKC activation (Fig. 5C, left) and brief enough to preserve cell viability (24). We found that single-cycle hypoxic preconditioning in glucose-free defined medium increased the proportion of surviving myocytes by 86% compared with control (66.4 Ϯ 1.9 versus 37.9 Ϯ 2.5%) following 9 h of hypoxia. Furthermore, the protective effect of single-cycle preconditioning was completely abolished by the ⑀PKC-selective antagonist ⑀V1-2 and unaffected by the control peptide ␤C2-4 (Fig. 5B). (Note that ␤C2-4 is a peptide derived from the ␤PKC C2 region that does not act as an ⑀PKC translocation inhibitor (21).) Therefore, ⑀PKC mediates, at least in part, the protective effect of hypoxic preconditioning in cultured cardiac myocytes. DISCUSSION In this study we describe a neonatal rat cardiac myocyte culture model in which two distinct hypoxic preconditioning protocols protected myocytes from death during a subsequent prolonged period of hypoxia. Two lines of evidence establish a central role for PKC in cardiac protection. First, hypoxic preconditioning selectively activated PKC isozymes. Second, direct activation of PKC with phorbol ester mimicked both the pattern of PKC isozyme activation and the protective effect of hypoxic preconditioning. These in vitro observations complement work described earlier with whole heart models in which PKC agonists mimicked the protective effect of ischemic preconditioning, and PKC antagonists inhibited myocardial protection (16,17). Our findings also support the hypothesis that components of the hypoxic preconditioning signal transduction pathway necessary for cardiac protection are present in ventricular myocyte culture models.
The principle finding of this study is that the role of a particular PKC isozyme in the preconditioning response can be probed with isozyme-selective PKC antagonists. Activation of PKC is associated with its translocation from the cell soluble to the particulate fraction (34) and with its binding to specific anchoring proteins within the particulate fraction termed RACKs (19,35,37). Activation of a particular PKC isozyme can be assayed by determining its relative distribution between cell fractions by Western blot analysis or by immunofluorescence staining. However, assignment of a biological function to a particular PKC isozyme on the basis of these techniques is limited because it is based on correlation. We have previously shown that transient permeabilization of the ⑀V1 fragment or ⑀V1-2 peptide into cardiac myocytes resulted in inhibition of PMA-induced ⑀PKC translocation without concurrent alteration of ␣-, ␤-, or ␦PKC translocation (20). Furthermore, PMAmediated negative chronotropy, known from earlier work to correlate closely with ⑀PKC translocation (14), was inhibited in cardiac myocytes permeabilized with the ⑀V1 fragment or ⑀V1-2 peptide but not with control peptide (20). In the present study ⑀V1-2 peptide inhibited both preconditioning induced translocation of ⑀PKC and protection of cardiac myocytes from hypoxic injury. In contrast, a selective translocation inhibitor of the C2-containing PKC isozymes did not affect hypoxic preconditioning, supporting a critical role for ⑀PKC in this biological function (Fig. 5B).
Transient permeabilization rather than transfection was chosen as the method of introduction of PKC inhibitor into cardiac myocytes because of well-documented transfection efficiencies of less than 5% in this culture model (38,39). For determination of changes in cell viability it is essential that the majority of myocytes contain the inhibitor. However, in a separate study we showed that stable expression of the ⑀V1 region in PC12 cells caused a similar selective inhibition of PMAinduced ⑀PKC translocation and function (36). As evidenced by Western blot analysis, expression of ⑀V1 fragment resulted in approximately 60% inhibition of ⑀PKC translocation following exposure to 30 nM PMA and no change in ␦PKC translocation. As a result of this isozyme-selective inhibition of PKC translocation, responses to nerve growth factors were altered (36). These data suggest that regardless of the method of introduction, these PKC antagonists selectively inhibit the translocation and function of their corresponding isozymes. Both ␦and ⑀PKC translocate in cardiac myocytes upon stimulation with PMA or following hypoxic preconditioning (Figs. [3][4][5]. However, because introduction of ⑀V1-2 abolished the protective effect of preconditioning during subsequent prolonged hypoxia (Fig. 5), we conclude that activation of ⑀PKC is necessary for PMA-induced and hypoxic preconditioning and predict that activation of ␦PKC is unlikely to mediate this acute form of myocyte protection. Instead, activation of ␦PKC may be an upstream event in the signal transduction pathway underlying the delayed form of myocyte tolerance to profound hypoxia known to occur 24 h after preconditioning (40). Alternatively, activation of ␦PKC may actually contribute to hypoxic injury. Studies in whole heart preparations suggest that PKC inhibition may at times reduce damage to cardiac tissue following severe ischemia (41). This controversy over the role of PKC in preconditioning may reflect opposing effects of ␦and FIG. 5. Activation of ⑀PKC is required for PMA-and hypoxic preconditioning-induced protection of cardiac myocytes from subsequent prolonged hypoxia. A, myocytes were permeabilized to introduce a peptide inhibitor of ⑀PKC (⑀V1-2), control peptide (⑀V1-3), or no peptide (-). Cells were then stimulated with 10 nM for 10 min (Phorbol Ester) or vehicle (Control), washed, and subjected to 9 h of hypoxia. 10 random fields per condition in each of four independent experiments were scored for cell viability. Preconditioning increased the proportion of surviving myocytes by 34% versus control. Protection was abolished by the ⑀PKC-selective antagonist ⑀V1-2. The number in each bar represents the total number of cells scored per condition. †, p Ͻ 0.05 versus control cells. *, p Ͻ 0.05 versus PMA-treated cells permeabilized in the absence of peptide (Ϫ) or in the presence of ⑀PKC-derived control peptide (⑀V1-3). B, myocytes were permeabilized to introduce ⑀V1-2, control (␤C2-4), or no peptide (-) and then exposed to 30 min of hypoxia in the absence of glucose (Preconditioned). After 30 min of recovery under normoxic conditions, preconditioned and control cells were subjected to 9 h of hypoxia. 20 random fields in each of two independent experiments were scored for cell viability. Preconditioning increased the proportion of surviving cells by 86% versus control. Protection was abolished by ⑀V1-2. The number in each bar represents the total number of cells scored per condition. † †, p Ͻ 0.05 versus control cells. **, p Ͻ 0.05 versus preconditioned cells permeabilized in the absence of peptide (-) or in the presence of control peptide (␤C2-4). C, left, myocytes were either maintained in glucose-supplemented medium under normoxic conditions (Nx) or incubated in glucose-free medium under hypoxic conditions (Hypoxia) for the times indicated. Western blot analysis revealed translocation of ⑀PKC from the soluble to the particulate fraction following 30 -120 min of hypoxia. Result is representative of two independent experiments. Densitometry revealed a 70% reduction in signal intensity at 120 min compared with normoxia in the soluble fraction and a 42% increase in the particulate fraction. C, right, myocytes were permeabilized in the presence of ⑀V1-2 (ϩ) or in the absence of peptide (Ϫ) and then either incubated under normoxic conditions (Nx) or exposed to 1 h of hypoxia (Hx) in glucose-free medium. Hypoxia induced translocation of ⑀PKC from the soluble to the particulate fraction that was inhibited by the ⑀PKC-selective peptide antagonist. Result is representative of two independent experiments. By densitometry, hypoxia caused reductions of 40 and 45%, respectively, in signal intensity in the second and third lanes of the soluble fraction compared with normoxia. The corresponding increase in the particulate fraction in the second lane was 47%, which was reduced by ⑀V1-2 to less than half (21% increase).
⑀PKC activation and emphasizes the importance of isozymeselective pharmacological tools in determining mechanisms of injury and protection. We have recently identified a ␦PKCselective peptide antagonist derived from the RACK-binding site on ␦PKC (20) which will be used to test these hypotheses regarding the role of ␦PKC activation in hypoxic preconditioning. Furthermore, studies with a novel ⑀PKC-selective agonist are underway, and preliminary data suggest that activation of ⑀PKC is sufficient to reduce hypoxia-induced cardiac myocyte death dramatically in this culture model.
The duration of ⑀PKC activation required for full protection from hypoxic cell death and the mechanisms underlying its down-regulation remain undefined. In addition, the stability of ⑀V1-2 in cells is unknown. We have previously shown that the initial intracellular concentration of peptide is approximately one-tenth that present in the extracellular permeabilization buffer (26). The marked potency of a potentially unstable peptide might be explained by protection from proteolysis following binding to ⑀PKC-specific RACKs. Alternatively, because ⑀PKC translocation in response to preconditioning does not persist (Fig. 4), even the transient presence of ⑀V1-2 peptide may have a profound effect on the downstream cellular events responsible for cardiac myocyte protection. Finally, although ⑀PKC mediates negative chronotropic effects in cardiac myocytes (14,20), it is not known whether slowing of contraction rate is necessary for protection from hypoxic injury. In a similar culture model, preconditioned myocytes ceased to contract earlier than paired control cells during a prolonged hypoxic challenge (18). However, the relationship between the earlier contractile failure of preconditioned myocytes and protection from hypoxic injury could not be determined from the available data.
In summary, we have shown that introduction of an ⑀PKC isozyme-selective antagonist into cultured cardiac myocytes inhibits the protective effects of preconditioning during subsequent prolonged hypoxia. These observations suggest that investigation of PKC isozyme-selective agonists and antagonists, or drugs that mimic the effects of these peptides, may result in new approaches to the treatment of cardiac and vascular disease.